Differences in Backbone Structure between ... - ACS Publications

Jul 1, 1995 - Graham J. Moore$,#. Department of Chemistry, University of Patras, Patras, Greece 261 10, Departments of Pharmacology and Therapeutics ...
0 downloads 0 Views 2MB Size
J . Med. Chem. 1995,38, 4660-4669

4660

Differences in Backbone Structure between Angiotensin I1 Agonists and Type I Antagonists John M. Matsoukas,*,’ George Agelis,’ Amal Wahhab,t John Hondrelis,’ Dimitris Panagiotopoulos,’ Raghav Yamdagni,o Qiao W U , Thomas ~ Mavromoustakos,LHernani L. S. Maia,” Renee Ganter,f and Graham J . Moore$,# Department of Chemistry, University of Patras, Patras, Greece 261 10, Departments of Pharmacology and Therapeutics and Chemistry, University of Calgary, Calgary Alberta, Canada T2N 4N1, National Hellenic Research Foundation, Institute of Organic and Pharmaceutical Chemistry, Athens 11 635, Greece, Department of Chemistry, University of Minho, Gualtar, P-4700 Braga, Portugal, and Department of Chemistry, University of Exeter, EX2 4QD Exeter, England Received March 28, 1995x

Type I angiotensin I1 antagonists with 0-methyl-L-homoserine [HSer(y-OMe)l and b-methoxyL-norvaline [Nva(G-OMe)l at position 8 have been prepared by the solid-phase method, purified by reverse-phase HPLC, and bioassayed in the rat uterus, and their backbone conformational properties were investigated by nuclear Overhauser effect (NOE) spectroscopy. [Sarl,HSer(y-OMe)81ANGII, [HSer(y-OMe)81ANGII, [Des1,HSer(y-OMe)81ANGII, [Sarl,N~a(d-OMe)~]the following antagonist activities, ANGII, and [ D ~ S ~ , N V ~ ( ~ - O M ~ ) ~had, I A Nrespectively, GII pA2: 7.6, 7.5, ~ 6 . 07.1, , and 6.9. Analogs of [Sar’IANGII with 6-hydroxy-L-norvaline [Nva(6OH)], 6-methoxy-L-norvaline [Nva(d-OMe)l, 4’-carboxyphenylalanine [Phe(4’-COOH)1, and 4‘(trifluoromethy1)phenylalanine [Phe(4’-CFs)l a t position 4 were also prepared by solid phase and bioassayed in the rat uterus. [ S ~ ~ ~ , N V ~ ( ~ - O H ) ~[Aib1,Nva(6-OMe)41ANGII, IANGII, [Sarl,DLhad, respectively, agonist activities Phe(4’-COOH)4]ANGII, and [S~~~,DL-P~~(~’-CF~)~]ANGII as follows: 4%, 1.5%, 3%, ,

, * m _ .

5

L I

6 1

5 5

5 ,

.ti,

,

4 6

,

, :
i

7,

7 ,

,

,

1 5

,.

1 ,

,

5

,

.

Figure 3. 'H NMR NOE difference spectra for [Sar',Nva(GOMe)81ANGII in DMSO-de.

Saturation of His C,H at 6-4.65 ppn

carried out in a similar way as for compound 3 using the method of Nagasawa.26 N-t-Boc-4-(trifluoromethyl)phenylalanine (10) was used for the synthesis of analog [ S~~~,DL-P~~(~'-CF~)~]ANGII, which was purified by HPLC identified by FAMBS and amino acid analysis and bioassayed in the rat uterus.

Discussion Structure-Activity Studies. [Sarl,IleslANGII (sarn i I I 1 ilesin) is a potent type I inhibitor CpA2 = 8.1) characterized by protracted antagonist effects on isolated smooth Reference Spectrum muscle tissues. Therefore, it was of interest to investigate structure-activity relationships in sarilesin analogs with modification a t positions l, 4, and 8 as well as conformational properties which could account for its antagonist activity. In the present study we have found that substitution of HSer(y-OMe) and Nva(d-OMe) a t position 8 in [SarlIANGII resulted in analogs with antagonist activities of pA2 = 7.6 and 7.1, respectively. These findings indicate that replacement of Iles in Figure 2. lH NMR NOE difference spectra for [HSer(y-OMe)*]Sarilesin with HSer(y-OMe) and its higher homolog ANGII in DMSO-ds. Nva(d-OMe)does not substantially change the bioactive conformation important for receptor recognition and phase methods. Deprotection of the Phe ring carboxyl expression of type I antagonist activity. However, the from its benzyl group took place simultaneously with lower antagonist activity of [Sarl,Nva(d-OMe)slANGII deprotection of other residues (Arg, His) during the bearing a longer aliphatic side chain at position 8 cleavage of the peptide from resin upon treatment with HF. The obtained analog, [Sar1,~~-Phe(4'-C00H)4]- suggests that the proper length of chain at the Cterminal residue of type I antagonist is important for a ANGII was purified by HPLC, identified by FABMS and favorable binding interaction. With the exception of amino acid analysis, and bioassayed in the rat uterus. [Desl,HSer(y-OMe)s]ANGII which was found to have Compound 10 was synthesized using diethyl acetavery low antagonist activity ( p A 2 < 61, variation in midomalonate and p(trifluoromethyllbenzy1 bromide as activity of the synthesized analogs illustrated a preferstarting materials in dry ethanol containing sodium ence for the lower homolog HSer(y-OMe) over Nva(dethoxide. The Gabriel condensation afforded compound OMe) for the expression of maximum type I antagonist 8 in high yield (91%). Hydrolysis of 8 with sodium activity. Replacement of Sar by Asp at position 1 in hydroxide followed by acidification afforded 4-(trifluo[Sarl,HSer(y-OMe)s]ANGII retained high antagonist romethy1)phenylalanine (9). Protection of the amino activity (pA2 = 7.5) while subsequent deletion of the group of 9 with the N-tert-butyloxycarbonyl group was

4664 Journal of Medicinal Chemistry, 1995, Vol. 38, No. 23

Matsouhas et al.

response. It has also been shown that replacement of N-terminal residue of position 1 resulted in loss of the Tyr hydroxyl by fluorine as in [Sar1,Phe(4'-Fl41antagonist activity (pAz < 6), indicating that the effect ANGII produces an antagonist (pA2 = 7.5)12 while of the positive charge at the N-terminus derives from polyfluorination of the Phe ring in [Sar1,Phe41ANGIIas its precise 10cation.~Substitution of Nva(6-OMe) in in [Sar1,Phe(Fj)4]ANGIIabolishes both agonist and position 4 of Sarilesin resulted in a drastic loss of antagonist activity, indicating that the affinity of ANGII activity (pAz -= 5) whereas for the analog [Aib',Nva(Ganalogs for receptors depends on the electronegative and OH)4,11eslANGII,a low binding affinity (pA2 = 6.2) was quadrupole properties of the aromatic ring.24 A recent observed, illustrating the importance of a free hydroxyl reportz5has provided evidence that the apparent bindgroup at position 4 for the expression of antagonist ing affinity of ANG I1 analogs is associated with a activity. perpendicular interaction of the position 4 aromatic ring Previous studies on ANGII have implicated that a with a receptor-based group. In this study, replacement triad of interacting groups (Tyr OH-His imidazole-Phe of the hydroxyl group by the trifluoromethyl group CF3 COO-) creates a tyrosinate anion pharmacophore (tyraproduced an analog without agonist or antagonist nophore) responsible for receptor activatiodtriggering. activity. The loss of activity illustrates that both This hypothesis is based on SAR, NMR, and fluoreselectrostatic and steric properties of the aromatic ring cence studies which further illustrated the importance in position 4 are subtle and highly relevant as suggested of the Tyr hydroxyl in receptor activation.15-18 In the previou~ly.~~~~~ present study, we have investigated the effects on bioactivity by substituting the Tyr residue in [SarllConformational Studies. Our recent investigations ANGII with other anionic or electronegative amino acids by NOE spectroscopy in the rotating frame (ROESY) as Nva(6-OH) and Nva(6-OMe) and with two aromatic and 1D-NOE studies of ANGII and [SarlIANGII have amino acids in which the Tyr hydroxyl has been shown proximity of the three rings (Tyr, His, Phe), a replaced by the COOH and CF3 groups, respectively. trans-His-Pro bond and a Tyr-Ile-His bend, which The phenylalanine derivatives, 4-(benzyloxycarbonyl)seems to be a critical factor for the manifestation of phenylalanine and 4-(trifluoromethyl)phenylalaninewere agonist activity.16-18 By saturating distinct resonances synthesized by novel methods. A variation of the at the well-resolved C, proton region in the NMR Gabriel synthesis utilizing diethyl acetamidospectra of the type I ANGII antagonists [HSer(y-OMe)81malonate20-22was used to prepare the (D,L) amino acids we were able to ANGII and [Sar1,Nva(6-OMe)81ANGII, 2 and 9 as depicted in Scheme 1. These amino acids shed light on the backbone conformation of these were incorporated in the [SarlIANGII molecule at molecules. Thus, saturation of the His C, proton position 4 in order to delineate aspects of this pharmaresonance in both antagonists did not produce any cophore which provides for its receptor triggering propenhancement of the Tyr C, proton resonance, in contrast erties. Deprotection of the 4-(benzyloxycarbonyl)to what was the case for ANGII and [ S ~ I - ~ I A N G I I . ~ ~ phenylalanine residue during treatment of the peptideSimilarly, saturation of the Tyr C, proton did not again resin with HF resulted in [Sar1,~,~-Phe(4'-COOH)lANGII produce any enhancement for the His C, proton. Lack which was found to be biologically inactive. Furtherof such enhancements in [HSer(y-OMe)81ANGII and more, the synthesis of 6-hydroxyl-L-norvaline [Nva(G[Sarl,N~a(6-0Me)~]ANG11 indicate a loose backbone OH)] was undertakenz3 in order to determine if an conformation a t this position of the molecules deprived aliphatic hydroxyl group appropriately positioned within of a Tyr-Ile-His bend, which is a characteristic feature the ANGII molecule would be able to mimic the tyrosine of ANGII, [SarlIANGII, and agonist analogs.ls The His hydroxyl group and participate in a charge relay sysC, and Tyr C, proton resonances constitute an excellent tem.17 The present study shows that [Sar',N~a(d-OH)~l- probe for investigating backbone conformation and their ANGII displays low agonist activity (4%) and presumchemical shifts are well-established at 6 = 4.65 ppm and ably has little tendency to participate in a stable charge 6 = 4.45 ppm for [HSer(y-OMe)81ANGIIand a t 6 = 4.63 relay system. Methylation of the hydroxyl group proppm and 6 = 4.48 ppm for [Sar1,Nva(6-OMe)81ANGII duces an analog, [ Sar1,Nva(6-OMe)4]ANGII,with simithrough COSY cross-peaks with their C,jp protons a t 6 lar activity (3%),suggesting that the electronegative = 2.90 ppml2.85 ppm and 6 = 2.88 ppml2.82 ppm, oxygen atom in both cases may act as a weak proton respectively. C l j protons ~ of aromatic residues (Tyr, His, acceptor and thereby produces a weak receptor response Phe) appear as distinct and easily recognizable AB (3-4%). quarters upon saturation of their adjacent C, protons when NMR experiments are carried out using DMSOWe thought it of interest t o investigate if a naked d a 2 0 (9/1)as solvent. Also, our previous NMR studies anion such as a carboxylate could act as a proton have constantly shown that the His C, and Tyr C, acceptor in place of the phenolate group of tyrosine in protons resonate at lower field than other C, protons, ANGII. Interestingly, [Sar1,~~-Phe(4'-COOH)4]ANGII and their saturation in our NOE studies provided useful was virtually biologically inactive, illustrating that the information concerning backbone conformation.l5-ls naked COO- anion does not contribute in the receptor Extension of these studies in sarilesin (unpublished activation process. Since both the carboxylate of Pheresults) supported our findings and indicated that a Tyr(4'-COOH) and the phenolate group of tyrosine at Ile-His bend as revealed by 1D-NOE studies is a position 4 can act as proton acceptors from the receptor, conformational feature of agonists but not of type I the vast difference in activity may indicate the imporantagonists. These findings may open new avenues for tance of having a nucleophilic group (rather than a designing non-peptide agonists as we have done with negative charge) provided by the tyrosinate species. The non-peptide ANGII receptor a n t a g o n i ~ t s . ~Our ~ - ~in~ nature of the aromatic ring of position 4 influences the vestigations of [HSer(y-OMeY%4NGII by 1D-NOE exbinding affinity of ANGII analogs and shows that periments have indicated also proximity of the Tyr and sufficient perturbation of the receptor by an aromatic His rings. Thus, saturation of the His Cd proton at 6 = ring can produce a partial response or eliminate the

ANGII Agonists and Type I Antagonists

Journal of Medicinal Chemistry, 1995, Vol. 38, No. 23 4665

Figure 4. Low-energy conformers of [SarlIANGII (top) and [HSer(y-OMe)s]ANGII(bottom).

6.88 ppm resulted in enhancement (7.1%) of the Tyr meta proton a t 6 = 7.02 ppm (Figure 2). Saturation of the Tyr ortho proton resonance at 6 = 6.60 ppm resulted in enhancement of the Tyr meta proton at 6 = 7.02 ppm but not of the His C2 and Cq proton resonances. These experiments show that the interaction between His and Tyr rings is not a reverse relaxation phenomenon. Thus, while the His Cz and Cq protons can relax through the closely spaced Tyr meta and ortho ring protons, the reverse effect is not observed upon saturation of the Tyr meta and ortho protons. The probable reason for this is that the Tyr ortho and meta protons have relaxation pathways which are not available to the His Cz and Cq protons. The Tyr ortho protons can relax through the Tyr meta and hydroxyl protons, while the meta protons can relax through the Tyr ortho and Cpp protons. In our studies, meta and ortho protons refer to the hydroxyl group of tyrosine. The NOE data suggest a loose backbone conformation for the type I ANGII antagonists [HSer(y-OMe)8]ANGIIand [Sar1,Nva(6-OMe)8]ANGIIin DMSO-&, in contrast to ANGII and [SarlIANGII which are characterized by a compact folded conformation due to the Tyr-Ile-His bend present in agonists.18 A comparison of the conformations of [SarlIANGII with [HSer(y-OMe)8]ANGII is shown in Figure 4. These models were derived using NOE data and represent low-energy structures. Details of constructing such models are given in a previous publication.18

Conclusion This research aims a t refining present knowledge of the angiotensin I1 backbone conformation as well as of the roles of residues 4 and 8 in activity and conformation of angiotensin I1 agonist and antagonist analogs. A TyrIle-His bend exists in ANGII and [SarlIANGII but not in type I antagonist analogs with an aliphatic amino acid at position 8. This conformational difference

between agonists and type I antagonists, as revealed by 1D-NOE studies, may constitute a major conformational feature distingushing the activity of agonists and antagonists and may provide new avenues in designing non-peptide ANGII agonists as has been possible for non-peptide ANGII antagonist^.^^-^^ A phenolate anion at position 4, but not a phenylalanine carboxylate anion, is required for the expression of agonist activity, suggesting a unique involvement of the tyrosine hydroxyl in receptor activation. Proper charge positioning or nucleophilicity is better provided by the tyrosinate species and accounts for its triggering ability. The SAFt data also indicate that for type I antagonists, substitution of Ile8 in Sarilesin with HSer(y-OMe) and Nva(6OMe) does not significantly change the active conformation important for receptor recognition and expression of antagonist activity. Variations in activity of the type I antagonist analogues illustrated a preference for Ile or Ala over HSer(y-OMe)and Nva(6-OMe). These data indicate that increasing the length of the position 8 aliphatic side chain results in reduction of antagonist activity and influences the conformation necessary for maximum interaction of a type I antagonist with the receptor.

Experimental Section N-t-BOGMethodology. Solid-phase synthesis of [Nva(GOH)4]-and [N~a(G-OMe)~]angiotensin I1 and Sarilesin analogs was accomplished using tert-butyloqcarbonyl-substituted amino acids and 1%cross-linked divinylbenzene resins (0.4-0.6 mequidg) as solid support and procedures previously described.1° N-t-Boc methodology was used also for the synthesis of [Phe(4'-COOH)4]- and [Phe(4'-CFd4]angiotensinI1 analogs. tert-Butyloxycarbonyl-blockedamino acids were synthesized in our laboratoriesz6 or were purchased from Peninsula Laboratories, Bachem Inc., or Chemical and Biopharmaceutical Laboratories of Patras (CBL). Protected amino acids, resins, ANGII, and sarilesin were purchased from Peninsula

4666 Journal of Medicinal Chemistry, 1995, Vol. 38, No. 23 Laboratories. Nva(d-OH) and Nva(d-OMe) were synthesised by methods previously described23or purchased from CBL and converted to N-t-Boc derivatives by the method of Nagasawa.26 Phe(4'XOOH) and Phe(4'-CFz) were synthesized by the Gabriel method described Solid-phase peptide synthesis was carried out with a Beckman 990 or 990B peptide synthesizer or with a reaction flask (2 x 12 cm) equipped with a bottom G-2 filter and tap connected to a water vacuum aspirator. Coupling steps with 2.5 equiv of Boc-protected amino acid and coupling reagent were employed for each amino acid; the coupling was mediated by Nfl-dicyclohexylcarbodiimide (DCC) in the presence of 1-hydroxybenzotriazole (HOBt) for 3 h. Deprotection steps (2 and 25 min) utilized 35% CF3COOH in CH2C12 containing indole (0.1%). Neutralization steps (2 and 25 min) utilized 10% triethylamine in CH2C12. Peptides were synthesized on a 0.5-mmol scale. The completed peptide was removed from the resin and simultaneously deprotected by treatment with anhydrous H F (10 mL) in the presence ofp-cresol(1 g) for 30 min a t 0 "C. The peptide was dissolved in CF3COOH (100 mL), and the resin was removed by filtration. The solvent was removed by rotary evaporation, and the crude product was isolated by trituration with ether and filtration. The crude product was dissolved in 7% acetic acid (10 mL), clarified by centrifugation, and purified by HPLC.

Matsouhas et al. acetate, giving 2 (16.66 g). A second crop (2.06 g) was obtained (89% total yield): 'H-NMR (D20) 6 7.85 (d, J = 8.2 Hz, 2H), 7 . 2 8 ( d , J = 8 . 2 H z , 2 H ) , 4 . 2 3 ( d d , J = 7 . 6 , 5 . 9 H z ,1H),3.2(m, 2H).

N - (tert-Butoxycarbonyl)-4-carboxyphenylalanine (3). tert-Butyl4,6-dimethylpyrimidine-2-thiolocarbonate (8.0 g, 32.3 mmol) was added t o a solution of amino acid 2

(7.434 g, 30.3 mmol) and Et3N (12.7 mL, 90.9 mmol) in tetrahydrofuran (68 mL) and H2O (68 mL). The mixture was stirred a t room temperature overnight according to the procedure of Nagasawa et aLz6 After workup 3 was obtained in 64% yield (5.945 g): 'H-NMR (CD30D) d 7.67 (d, J = 8.2 Hz, 2H), 7.07 (d, J = 8.2 Hz, 2H), 4.17 (dd, J = 9.1, 5.4 Hz, lH), 2.82 (two sets of dd overlapping, 2H), 1.47 (s, 9H); NMR (DMSO-ds)6 173.2 (COzH), 167.2 (COzH), 155.3 (CON), 143.3, 129.2, 129.1, 128.9, 78.0 (CO), 54.7 (CHNj, 36.4 (CHzPh), 28.1 (t-Bu); exact mass calcd for CljH20NOj (M' - OH) 292.1185, found 292.1191. N - (tert-Butoxycarbonyl)-4-carboxyphenylalanine Dibenzyl Ester (4). Dicyclohexylcarbodiimide (9.374 g, 45.4 mmol) was added to a mixture of 3 (6.692 g, 21.6 mmol), benzyl alcohol (4.913 g, 45.4 mmol), and DMAP (0.55 g) in CHzClz (100 mL). After the reaction mixture was stirred overnight, the precipitated dicyclohexylurea was filtered out and rinsed with the minimum amount of CHzClz. Workup according to Synthesis of 4-Carboxy- and 4-(Trifluoromethyl)the procedure of Hassner and Alexanianz2 gave crude 4. phenylalanine. A variation of the Gabriel synthesis utilizing Chromatography over flash silica eluting with 0-2% MeOHdiethyl acetamidomalonate2' was used to prepare the (D,L) CHC13 gave 4 in 63% yield (6.7 g): 'H-NMR (CDC13)6 7.94 (d, amino acids 2 and 9 as depicted in Scheme 1. Selective J = 8.2 Hz, 2H), 7.50-7.25 (m, lOH), 7.11 (d, J = 8.2 Hz, 2H), deprotection of the benzyl ester 5 was achieved by the copper 5.37 (s, CH20, 2H), 5.13 (d, J = 1.2 Hz, CH20, 2H), 4.66 (m, complex according to the method of Prestidge et a1.22 Amino l H ) , 3.15 (m, CHZPh, 2H), 1.43 (s, t-Bu, 9H); T - N M R (CDC13) acid 7 was obtained in 90% purity by 'H-NMR with the d 171.3 (Con), 166.1 (COS),163.2 (CON), 141.5, 136.0, 134.9, impurity tentatively identified as the structural isomer Boc129.8, 129.4, 129.3, 128.8, 128.6, 128.2, 128.1, 80.0 (t-BuCO), NH-CH-(CH2Ph-4-C02H)-COzCHzPh. The Boc-N-protected 67.2 (CHzO), 66.6 (CHZO), 54.2 (CHN), 38.3 (CHzPh),28.2 (tamino acids 7 and 10 were used in position 4 to prepare ANGII Bu); exact mass calcd for C ~ ~ H ~ I N (M+ OE - C4H100)415.1420, analogs. Diethyl 2-Acetamido-2-[[4-(ethoxycarbonyl)phenyll- found 415.1434. 4-CarboxyphenylalanineDibenzyl Ester (5). Trifluomethyllmalonate (1). Dry ethanol (68 mL) was added roacetic acid (50 mLj was added to a solution of 4 (6.7 g) in dropwise to pentane-washed sodium pieces (2.3 g, 0.1 mol) CH2C12 (50 mL) a t room temperature, and the progress of the under an argon atmosphere. After completion of addition, reaction was followed by TLC. After 40 min, excess TFA and heating was required to dissolve the last pieces of sodium. Solid CHzClz were removed in uacuo. Ethyl acetate (100 mL) was diethyl acetamidomalonate 121.722 g, 0.1 mol) was added then added, and the organic layer was washed with H20 (2 x portionwise over 30 min, after which the reaction mixture 50 mLj, 5% NaHC03 (3 x 50 mL), H20 (50 mL), and brine (50 became very thick and hard to stir and heating was required mL). Finally, it was dried over anhydrous MgS04, filtered, to solubilize it. a-Bromo-p-toluic acid ethyl ester (24.31 g, 0.1 and evaporated, giving 5 in 98% yield (5.29 g): 'H-NMR mol, 70% pure, the balance being p-toluic acid ethyl ester (CDC13)d 8.06 (bs),7.81 (d, J = 8.3 Hz, 2H), 7.48-7.0 (m, 12H), 16.7% and a,a-dibromo-p-toluic acid ethyl ester) in dry ethanol 5.22 (s, CHzO), 5.0 (d, J = 2.8 Hz, CH20, 2H), 4.26 (m, l H ) , (100 mL) was added dropwise over 90 mimzO The mixture 3.19 (d, J = 6.3 Hz, 2H); I3C-NMR(CDC13)d 172.3 (Cos), 166.1 turned yellow, and a fine yellowish precipitate deposited on (COz), 141.5, 136.0, 134.9, 130.0, 129.4, 129.0, 128.5, 128.2, the sides of the flask. After the mixture was refluxed for 5 h, 128.1, 67.3 (CHzO), 66.6 (CHzO), 55.0 (CHN), 39.4 (CHzPh); it was cooled, ether and H20 were added, and the organic layer exact mass calcd for Cz4H23N04(M+)389.1627, found 389.1624. was separated. The aqueous layer was saturated with solid NaCl and extracted with ethyl acetate. The organic extracts 4-(Benzyloxycarbonyl)phenylalanine(6).A solution of CuSO4*5H20(16.157 g) in H2O (283 mL) was added to 5 (5.3 were combined, washed with brine, dried over MgS04, and filtered. The solvents were removed in uacuo. Hexaneg, 13.6 mmol) in ethanol (113 mL), and the pH was adjusted to 8 with 1 N NaOH according to the procedure of Prestidge benzene, 4:1, was added to the residue, and the insoluble et ~ 1 . ~After ' the mixture was stirred a t 33 "C for 1 h, the pH material was filtered out and washed exhaustively with 1O:l was lowered to 3 and the insoluble blue copper complex was hexane-benzene. Then, it was allowed to air-dry giving 1 filtered out and washed with HzO, ethanol, ethyl acetate, and (23.4 g). A second crop (9.0 g) could be obtained after ether. Ethylenediaminotetraacetic acid disodium salt (6.32 g, concentration of the mother liquor (85% total yield): 'H-NMR 17 mmol) in water (81 mL) was added, and the solution was (CDC13) d 7.97-7.91 (m, 2H), 7.1-7.06 (m, 2H), 6.53 (bs, NH, boiled. On cooling, a white solid precipitated. This was lH), 4.41-4.42 (overlapped quartets, OCH2,6H), 3.71 (s, CHzfiltered and washed with water and was allowed to air-dry, Ph, 2H), 2.03 (s, COCH3, 3H), 1.41-1.26 (overlapped triplets, 9H); I3C-NMR (CDC13) d 169.1, 167.3, 166.3, 140.6, 129.8, giving 6 in 83% yield (3.4 g). 129.7, 129.5, 67.0,62.8,60.9,37.8,23.0,14.3, 14.0; exact mass N-(tert-Butoxycarbonyl)-4-(benzyloxycarbonyl)calcd for C19HzsN07 379.1631, found 379.1637. phenylalanine (7). Amino acid 6 (1.4 g, 4.7 mmol) was stirred with 2-[[(tert-butoxycarbonyl)oxyliminol-2-phenyl4-Carboxyphenylalanine Hydrochloride (2). Sodium acetonitrile (1.264 g, 5.1 mmol) and Et3N (1.8 mL) in dioxane hydroxide (11.6 g, 0.29 mol) in HzO (226 mL) was added to (10 mL) and HzO (10m) at room temperature overnight. After acetamido triester 1 (32.4 g, 85.4 mmol) in tetrahydrofuran workup, the crude product was triturated from 2:8 mixture of (80 mL), and the mixture was stirred at room temperature ethyl acetatehexane, giving 7 in 85% yield (1.58 g). The IHovernight. After the bulk of tetrahydrofuran has been removed NMR spectrum showed the product to contain a 10% impurity, in uacuo, acetic acid (24.5mL) was added to acidify the reaction tentatively identified as an isomer of 7 in which the benzyl and the mixture was refluxed for 1 h. Then the mixture was protecting group is on the a-carboxy group: 'H-NMR (CDCl3) taken to dryness, 12 M hydrochloric acid (80 mL) was added, 6 8.15 (bs), 8.02 (d, J = 8.2 Hz, 2H), 7.45-7.25 (m, 7H), 5.36 and the mixture was refluxed for 1h. The mixture was taken ( s , CHz0, 2H), 4.65 (m, lH), 3.2 (m, 2H), 1.41 (s, t-Bu, 9H); to dryness, Hz0 was added, and it was taken to dryness again. I3C-NMR (CDC13) d 175.9 (COz), 166.2 (COa), 155.7 (CON), Water was added to the residue, and the insoluble material 142.0, 136.0, 130.0, 129.5, 129.0,128.6, 128.2, 128.1,80.1 (CO), was filtered out and washed with HzO, ethanol, and ethyl

AiVGII Agonists and Type I Antagonists

Journal of Medicinal Chemistry, 1995, Vol. 38, No. 23 4667

66.7 (CHzPh),28.2 (t-Bu);exact mass calcd for CzzHzbNOs (M+) 399.1682, found 399.1686. Diethyl 2-Acetamido-2[[4-(trifluoromethyl)phenyl]methyllmalonate (8). p(Trifluoromethy1)benzyl bromide (2.0 g, 8.37 mmol) in dry ethanol (8.5 mL) was added dropwise over 30 min under argon to a solution of diethyl acetamidomalonate ester (1.817 g, 8.37 mmol) and sodium ethoxide in dry ethanol [prepared in situ by the reaction of sodium metal (0.192 g, 8.37 mmol) with ethanol (6.0 mL)]. After completion of addition, the mixture was refluxed for 5I/2 h, and then it was stirred a t room temperature for 12 h. Water was added to dissolve the yellow precipitate, and the mixture was extracted with chloroform (3x), The combined organic extracts were washed with brine, dried over magnesium sulfate, and filtered, and the solvent was removed in vacuo. Chromatography of the residue over flash silica, eluting with 24-32% ethyl acetatehexane, gave 8 in 91% yield (2.86 8): ‘H-NMR (CDC13)6 7.53 (d, J = 8.1 Hz, 2H), 7.13 (d, J = 8.1 Hz, 2H), 6.54 (bs, NH, lH), 4.28 (q, J = 7.2 Hz, CH20,4H),3.73 ( S CH2Ph, 2H), 2.03 ( s , CHzCO, 3H), 1.30 (t, J = 7.1 Hz, 6H). Anal. Calcd for C17H20F3N05:C, 54.40; H, 5.37; N, 3.73. Found: C, 54.4; H, 5.26; N, 3.74. 4-(Trifluoromethy1)phenylalanineHydrochloride (9). The diester 8 (2.76 g, 7.35 mmol) was dissolved in tetrahydrofuran (10.0 mL) and water (13.0 mL). Solid sodium hydroxide (0.67 g, 16.5 mmol) was then added, and the mixture was stirred a t room temperature for 18 h. Glacial acetic acid (2.0 mL) was added, and the mixture was boiled for 1 h. It was then concentrated under reduced pressure. Concentrated hydrochloric acid (10 mL) was added, and the resulting slurry was refluxed for 1h. The mixture was taken t o dryness, and the residue was taken up into ethanol. The ethanol was removed in uacuo. The remaining residue was refluxed once again with concentrated hydrochloric acid (10 mL) for 1h, and then it was taken to dryness. Water was added to the residue, and the insoluble material was filtered out, washed with ether, and air-dried, giving 9 (1.0 g). The filtrate was taken t o dryness, and dry ethanol was added. The insoluble material was filtered out. After evaporation of the ethanol filtrate, the residue was triturated with ether, giving a second crop of 9 (0.819 g). The hydrochloride 9 was obtained in 92% total yield: ‘H-NMR (CD30D) 6 7.67 (d, J = 8.1 Hz, 2H), 7.51 (d, J = 8.1 Hz, 2H), 4.33 (dd, J = 6.0, 7.3 Hz, CHC02, l H ) , 3.433.20 (m, CHzPh, 2H), 13C-NMR(CD30D) 6 170.9 (COzH),140.4, 131.3, 130.9, 127.0, 126.93, 126.85, 54.8 (CHN), 37.0 (CH2Ph). Anal. Calcd for C10HIoF3N02: C, 44.54; H, 4.11; N, 5.19. Found: C, 44.23; H, 3.95; N, 5.02. N-t-BOC-4-(trifluoromethyl)phenylalanine (10). To a solution of hydrochloride 9 (1.771g, 6.57 mmol) and triethylamine (2.2 mL, 16.4 mmol) in dioxane (12.0 mL) and water (12.0 mL) was added 2-[[(tert-butoxycarbonyl)oxyliminol-2phenylacetonitrile (BOC-ON, 1.643 g, 6.67 mmol), and the reaction mixture was stirred at room temperature for 21 h. Dioxane was removed in uacuo, and the mixture was diluted with water (30 mL). The basic aqueous layer was extracted with ether (3 x 50 mL). The aqueous layer was acidified with 5 N hydrochloric acid to pH 3, saturated with sodium chloride, and extracted with ethyl acetate (4 x 50 mL). The combined ethyl acetate extracts were dried over magnesium sulfate, filtered, and concentrated in uacuo. Chromatography of the residue on a flash silica column eluting with 4-7% MeOW CHC13 gave 10 in 67% yield (1.46 g): IH-NMR (CDC13 2 drops of CD30D) 6 173.8 (C02H), 155.3 (NCO), 140.5, 129.8, 128.9, 125.22, 80.0 (CO), 54.0 (NCH), 37.8 (CHnPh), 28.1 ((CHds). Anal. Calcd for CljH18F~N04:C, 54.05; H, 5.44; N, 4.20. Found: C, 53.86; H, 5.34, N, 4.21. Fmoc Methodology. For the synthesis of [HSer(y-OMe)*] and [Nva(6-OMe)*]angiotensin I1 analogs we resorted to the Fmoc methodology utilizing the 2-chlorotrityl-resin as solid support as reported by Barlos et al.27 Fmoc-blocked amino acids were synthesized in our laboratories as described previously or were purchased from CBL. Fmoc-His(Trt)-OH and Fmoc-Arg (Pmc)-OH were purchased from CBL and NOVA BIOCHEM. Attachment of the first Fmoc-protected amino acid, HSer(y-OMe) or NVa(G-OMe),to the resin (2-chlorotrityl) was achieved by a simple, fast, and racemization-free reaction using diisopropylethylamine (DIPEA) in dichloromethane

(DCM) solution a t room t e m p e r a t ~ r e .Solid-phase ~~ peptide syntheses were carried out on 2-chlorotriphenylmethyl (trityl) resin (1.4-1.6 mmol Cl-/g) using a manually handled reaction vessel (2 x 12 cm) equipped with a porous G filter (size 2) and tap a t the bottom connected with a water vacuum aspirator. A vibrator was used for shaking the reaction vessel throughout the several steps. The protocol used for the peptide synthesis is previously des~ribed.~’ After preactivation with l-hydroxybenzotriazole (HOBt) and dicyclohexylcarbodiimide (DCC), a 2.5-fold molar excess of Fmoc amino acids were used. Deprotection steps (5 and 20 min) utilized 20% piperidine in DMF. Splitting of the peptide from the 2-chlorotrityl-resin was carried out by treating the peptide-resin with the splitting solution, dichloromethane-acetic acid-2,2,2-trifluoroethanol (7:1:2, 15 mL/g resin) for 1 h. The mixture was filtered off, and the resin was washed with the splitting mixture and DCM several times. The solvent was removed on a rotary evaporator, and the obtained oily product was treated for 45 min a t room temperature with 45-50% trifluoroacetic acid in DCM, containing 5% anisole as scavenger. The mixture was evaporated to -0.5 mL, and the peptide precipitated from dry diethyl ether as a white amorphous solid. Removal of the protecting groups (Sar-Boc,Arg-Pmc, His-Trt) was achieved by treatment of the protected peptide with TFA-CHzClz ( l : l , 20 m u g peptide) in the presence of scavenger (5% anisole). The solvent was removed by rotary evaporator, and the crude product was isolated by trituration with ether and filtration. The crude product was dissolved in 7% acetic acid (10 mL), clarified by centrifugation, and purified by HPLC. Preparative Reversed-Phase HPLC. Purification of peptides was accomplished with use of a Varian HPLC system equipped with a Vista 401 microprocessor controller. Separations were achieved on a BioRad Hi-Pore 318 reversed-phase preparative column (25.0 x 2.15 cm) a t 25 “C with a stepped linear gradient of acetonitrile in 0.1% CF3COzH at a flow rate of 7.5 mL/min. Automated repetitive injections of peptides (5 x 5 mg) were made from a nitrogen-pressurized Rheodyne injector with a 2.0 mL sample loop. One-fifth of the total sample was injected during each run by lowering the flow rate to 4.0 m u m i n for 0.1-min “inject”period. One cycle consisted of the following events: 0-10 min, 7.5 m u m i n , 90% H2010% of 1%aqueous CF3C02H; 10-11 min, 4.0 m u m i n ; 1111.1 min, “inject”; 11.1-13 min, 7.5 m u m i n , 70% H20-20% CH3CN-10% of 1%CF3C02H; 13-30 min, 45% H20-45% CH&N-lO% of 1%CF3C02H;30-42 min, 90% CHzCN-lO% of 1% CF3C02H; 42-50 min, 100% H20. Fractions were collected a t 0.1-min intervals with a Gilson Model 210 fraction collector programmed t o collect for a 5-min period centered around the elution time (20-25 min) ofthe major product. The fraction collector was restarted by the Vista 401 a t the beginning of each HPLC run so that the material eluted with the same retention time was repeatedly deposited in the same tubes. Elution of the peptide was detected simultaneously from the absorbances a t 254 nm (Varian UV-1) and 230 nm (Kratos SF769Z). The elution times for these peptides were similar (22-24 min). Fractions containing the major product were pooled; after removal of CH3CN on a rotary evaporator a t 40 “C, the fractions were lyophilized and stored at -20 “C, yield 5-20%; purity ’98%. Amino acid analyses (Beckman 121M) were carried out after acid hydrolysis in 6 N HC1 (containing 1%cresol t o prevent loss of tyrosine) a t 100 “C for 18 h in uacuo. Thin-Layer Chromatography (TLC). TLC was carried out with precoated silica gel on glass (Merck Kiesel gel 60 F254) TLC plates. The two-solvent systems used were l-butanol-pyridine-acetic acid-water (15:10:3:6, BPAW) and chloroform-methanol-acetic acid-water (15:10:2:3,CMAW). Peptides were detected by fluorescence with ninhydrin (0.3%) in 1-butanol-acetic acid (100:3, v/v). Rat Uterus Bioassay. The rat isolated uterus was performed as described below. Female virgin SpragueDawley rats (150-200 g), injected 24 h previously with 25 pg of diethylstilbestrol in 0.05 mL of 20% polyvinylpyrrolidone (im), were killed by stunning and exsanguination. The uterine horns were exposed, cleaned of adherent fat in situ, and then removed t o a Petri dish containing DeJalons solution (150 mM NaC1, 5.6 mM KC1, 0.18 mM CaC12, 1.8 mL NaHC03 and 1.4

+

4668 Journal of Medicinal Chemistry, 1995, Vol. 38, No. 23 mM glucose, PH 7.0) a t room temperature where each was divided into two half lengths and arranged for mounting in a 3 mL tissue bath. Tissues thus prepared were tied to Grass force-displacement transducers under 1 g of tension and bathed a t 29 "C in a constantly oxygenated DeJalons solution as prepared above. Contractions were monitored with Gould Metripak 763341-4202 isotonic transducers coupled to Gould 13-4615-50 transducer amplifiers housed in a Gould 2600s recorder. Samples were prepared initially as a M solution in water. Agonist activities of peptides were determined by matching the response with an equivalent response to ANGII (human). For the measurement of antagonist activity of peptide ANGII receptor antagonists the analogs were given 2 min before ANGII. Antagonist activities (pA2) were determined as the negative logarithm of the concentration of antagonist required to reduce the response to a n EDSOdose of ANGII to that of a n ED502 dose. Each analog was tested in a t least three different tissues, and the mean result was taken. A noncumulative dose-response curve to ANGII was first established in each tissue, and the ECja dose and ECjO2 response was determined. To avoid tachyphylaxis, the tissue was washed out a t the peak response to the agonist and was allowed to rest for 10 min between agonist challenges.8,2e Antagonist was applied to the tissue 2 min before the agonist, and the reproducibility of the ECso agonist response was established before and after each antagonist challenge. NMR Experiment. NMR experiments were carried out using a Bruker 400-MHz NMR spectrometer. [HSer(y-OMe)18]ANGII and [Sar1,Nva(b-OMe)81ANGIIwere studied by dissolving three milligrams of each in 0.5 mL of DMSO-d6. The chemical shifts were reported relative t o the undeuterated fraction of the methyl group of DMSO-ds at 2.50 ppm with respect to TMS. One-dimensional spectra were recorded with a sweep width of 4.500 Hz and 32K (zero filled to 64K) data points and by methods previously described.16 The COSY (twodimensional correlated spectroscopy) experiment provided contour plots which were symmetrized with respect to the diagonal. One-dimensional NOE experiments were carried out in the difference mode using multiple irradiation. The methods used for the COSY and NOE experiments were similar to those previously d e s ~ r i b e d . ~ ~ . ' ~ ~ ~ " ~ ~ ~ ~ ~

Acknowledgment. This work was supported by European Community (EC) grants (BIOMED programme No. 920038, BIOMED-PECO No. 930158, and COPERNICUS No. 9402381, the Ministry of Energy and Technology of Greece, the Alberta Heart Foundation, and Bristol Myers-Squibb. We also acknowledge NATO support (Grant CRG 940612). We thank Mr. Dimitris Ballas for typing the manuscript. References Page, I. H. Hypertension Mechanisms; Harcourt Brace Jovanovich: New York, 1987; pp 347-470. ( 2 ) Khosla, M. C.; Smeby, R. R.; Bumpus, F. M. Structure-activity relationship in aneiotensin I1 analogs. In Handbook of ExDerzmental Pharmac&gy; Page, I. H . , Bumpus, F. M., Eds.; Springer-Verlag: h'ew York, 1974; Vol. 37, pp 126-161. Turker, R. K.; Hall, M. M.; Yamamoto, M.; Sweet, C. S.; Bumpus, F. M. New, long-lasting competitive inhibitor of angiotensin. Science 1972,177, 1203-1205. Khosla, M. C.; Leese, R. A,; Maloy, W. L.; Ferreira, A. T.; Smeby, R. R.; Bumpus, F. M. Synthesis of some Analogs of Angiotensin 11 as Specific Antagonists of the Parent Hormone. J . Med. Chem. 1972,15, 792-795. Sen, I.; Jim. K. F.; Soffer, R. I. Solubilization and characterization of a n angiotensin I1 binding protein from liver. Eur. J . Biochem. 1983,136, 41-49. Pals, D. T.; Masucci. F. D.: Denniro, G. S.. J r . ; Sipos, F.; Fessler, D. C. Role of the Pressor Action of Angiotensin I1 in Experimental Hypertension, Cir. Res. 1971,29. 673-681. Munoz-Ramirez, M. H.; Khosla, M. C.; Hall, M. M.; Bumpus, F. M.; Khairallah. P. A. In citro and in ciuo Studies of [l-Sarcosine, 8 Threonine] Angiotensin 11. Res. Commun. Chem. Path Pharmacol. 1976,13, 649-663. Scanlon, M. N.; Matsoukas, J. M.; Franklin, K. J.; Moore, G. J. A new approach t o angiotensin I1 antagonists: Methylation of the tyrosine hydroxyl in ANG 11. Life Sei. 1984,34. 317-321. i 1)

Matsoukas et al. ( 9 ) ( a )Matsoukas, J . M.; Cordopatis, P.; Belte, U.; Goghari, M. H.; Ganter, R. C.; Franklin, K. J.; Moore, G. J. Importance of the N-terminal domain of the type I1 angiotensin antagonist Sarmesin for receptor blockade. J. Med. Chem. 1988,31, 1418-1421. ib) Moore, G. J.; Ganter, R. C.; Goghari, M. H.; Franklin, K. J. Importance of the h'-terminal domain of the type I angiotensin I1 antagonist [Sar', Ilea]ANG I1 for receptor blockade. Int. J . Pept. Protein Res. 1991,38,1-7. (101 Matsoukas, J. M.; Agelis, G.; Hondrelis, J.; Yamdagni, R.; Wu, Q.; Ganter. R.; Smith. J. R.; Moore, D.; Moore. G. J. Synthesis and Biological Activities of Angiotensin 11, Sarilesin and Sarmesin Analogues containing Aze or Pip a t position 7. J . Med. Chem. 1993,36. 904-911. I 111 Matsoukas, J. M.; Goghari, M. A,; Scanlon, M. N.; Franklin, K. J.;Moore, G. J. Synthesis and biological activities of analogues of angiotensin I1 and 111 containing 0-methyltyrosine and D-tryptophan. J . Med. Chem. 1985,28,780-783. 112, Samanen, J.; Cash, T.; Narindray, D.; Brandeis, E.; Yellin, T.; Regoli. D. The Role of Position 4 in Angiotensin I1 Antagonism: A Structure - Activity Study. J . Med. Chem. 1989,32, 13661370. (13, Bovy. P. R.; O'Neal, J. M.; Olins, G. M.; Patton, D. R.: McMahon, E. G.: Palomo, M.; Koepke, J . P.; Salles, K. S.; Trapani, A. J.; Smits. G. J.; McGraw, D. E.; Hutton, W. C. Structure-Activity Relationships for the Carboxy-Terminus Truncated Analogues of Angiotensin 11. a New Class of Angiotensin I1 antagonists. J . Med. Chem. 1990.33,1477-1482. 1141( a ) Moore, G. J. Kinetics of acetylation-deacetylation of angiotensin 11. Intramolecular interactions of the tyrosine and histidine side chains. Int. J . Pept. Protein Res. 1985,26,469481. ( b )Moore, G. J.: Matsoukas, J . M. Angiotensin as a model for hormone-receptor interactions. Biosci.Rep. 1985,5, 407416. ( 1 5 ) Matsoukas, J. M.; Bigam, G.; Zhou, N.; Moore, G. J. 'H-NMR studies of ISar'lAngiotensin I1 conformation by nuclear Overhauser effect spectroscopy in the rotating frame (ROESY): Clustering of the aromatic rings in dimethylsulfoxide. Peptides 1990,11, 359-366. 116) Matsoukas, J. M.: Yamdagni, R.; Moore, G. J . 'H-NMR studies of Sarmesin and [Des'] Sarmesin conformation in dimethylsulfoxide by nuclear Overhauser effect (XOEi enhancement spectroscopy: Folding of the N- and C-terminal domains. Peptides 1990,1 1 , 367-374. I 17, Turner, R. J.: Matsoukas, J. M.; Moore, G. J. Fluorescence properties of angiotensin I1 analogues in receptor-simulating environments: relationship between tyrosinate fluorescence lifetime and biological activity. Biochim. Biophys.Acta 1991, 1065,21-28. 118 I Matsoukas, J. M.; Hondrelis, J.;Keramida, M.; Mavromoustakos. T.; Makriyannnis, A,; Yamdagni, R.; Wu, Q.; Moore, G. J . Role of the N-terminal Domain of ANGII and [Sar'lANGII on conformation and activity: NMR Evidence for Aromatic Ring Clustering and Peptide Backbone Folding compared to Ah'GII. J . Biol. Chem. 1994,269,5303-5312. 119) Moore, G. J.; Franklin, K. J.; Nystrom, D. N.; Goghari, M. H. Structure-desensitization relationships of angiotensin analogues in the rat isolated uterus. Can. J . Physiol.Pharmacol. 1985, 63, 966-971. ( 2 0 ) Vogel, A. Text book ofpractical Organic Chemistry, 4th ed.; Longman: London, 1978; p 554. 1211 Prestidge. R. L.; Harding, D. R. K.; Battersby, J. E.; Hancock, U'.S. A novel High-Yield Synthesis of 7-Esters of Glutamic Acid and /%Esters of Aspartic Acid by the Copper-Catalyzed Hydrolysis of Their Diesters. J . Org. Chem. 1975,40,3287-3288. (221 Hassner, H.; Alexanian. V. Direct room temperature Esterification of carboxylic acids. Tetrahedron Lett. 1978,4475-4478. 123) Barlos. K.; Mamos, P.; Papaioannou, D.; Patrianakou, S. Simple Syntheses of is,-2- and 4 Amino-5-hydroxypentanoic Acids. Chena. Commun. 1987,1583-1584. 124) ( a ) Guillemette, G.; Bernier, M.; Parent, P.; Leduc, R.; Escher, E. Angiotensin 11: Dependence of Hormone Affinity on the Electronegativity of a Single Side chain. J . Med. Chem. 1984, 27, 315-320. ib) Fowler, P. J.: Moore, G. J. Calculation of the magnitude and orientation of electrostatic interactions between small aromatic rings in peptides and proteins: Implications for angiotensin 11. Biochem. Biophys. Res. Commun. 1988,153, 1296-1300. ic) Bovy. P. R.; Getman, D. P.; Matsoukas, J. M.; Moore, G. J. Influence of polyfluorination of the phenylalanine ring of angiotensin I1 on conformation and biological activity. Biochim. Biophys. Acta 1991,1079, 23-28. (25) Moore, G. J.; Ganter, R. C.; Matsoukas, J. M.; Hondrelis, J.; Agelis, G.: Barlos. K.; Wilkinson, S.; Sandall, J.; Fowler, P. Receptor interactions of the position 4 side chain of angiotensin I1 analogues: Importance of aromatic ring quadrupole. J . Mol. Rec. 1994,7 , 251-256. 126) Nagasawa, T.; Kuroiwa, K.; Narita, K.; Isowa, Y. New agents for t-butyloxycarbonylation and p-methoxybenzyloxycarbonylation of amino acids. Bull. Chem. Soc. Jpn. 1973,46, 1269-1272.

ANGII Agonists and Type I Antagonists Barlos, K.; Gatos, D.; Hondrelis, J.; Matsoukas, J. M.; Moore, G. J.; Schafer, W.; Sotiriou, P. Preparation of New Acid-Labile Resins of secondary alcohol type and their application in peptide synthesis. Liebigs Ann. Chem. 1989,951-955. Freer, R. J . ; Stewart, J. M. Synthesis and Pharmacology of position 6 analogs of angiotensin 11. J . Med. Chem. 1973,16, 733-735. Hogg, A. M. Conversion of mass spectrometers for fast-atom bombardment using easily constructed components. Int. J . Mass Spectrom. Ion Phys. 1983,49, 25-34. Noggle, J. H.; Shirmer, B. E. The Nuclear Overhauser Effect; Academic Press: New York, 1971. Marion, D.; Wiithrich, K.; Application of phase sensitive twodimensional correlated spectroscopy (COSY) for measurement of 'HJHspin-spin coupling constants in proteins. Biochem. Biophys. Res. Commun. 1983,113, 967-974. Wahhab, A,; Smith, J.; Moore, G. J. Synthesis and Biological Activities of Non-peptide Angiotensin Receptor Antagonists with Tetrazole Replaced Imidazole Ring. Drug Res. 1993,43, 828830. Wahhab, A,; Smith, J . R.; Ganter, R. C.; Moore, D. M.; Hondrelis, J.; Matsoukas, J. M.; Moore, G. J . Imidazole Based Non-Peptide

Journal of Medicinal Chemistry, 1995, Vol. 38, No. 23 4669 Angiotensin I1 Receptor Antagonists: An investigation of the Effect of the Orientation of the Imidazole Ring on Biological Activity. Drug Res. 1993,4 3 , 1157-1168. (34) Smith, J. R.; Wahhab, A,; Moore, D.; Ganter, R. C.; Moore, G. J. Transposition of the Imidazole Ring Substituents of Angiotensin Mimetics Related to Losartan. Drug Discovery Des. 1994,12, 113-119. (35) Duncia, J. V.; Chiu, A. T.; Carini, D. J.; Gregory, G. B.; Johnson, A. L.; Price, W. A,; Wells, G. J.; Wong, P. C.; Calabrese, J. C.; Timmermans, P. B. M. W. M. The discovery of Potent Nonpeptide Angiotensin I1 Receptor Antagonists: A New Class of Potent Antihypertensives. J . Med. Chem. 1990,33, 1312-1329. (36) Carini, D. J.;Duncia, J.V.; Aldrich, P. E.; Chiu, A. T.; Johnson, A. L.; Pierce, M. E.; Price, W. A,; Santella, J. B., 111; Wells, G. J.; Wexler, R. R.; Wong, P. C.; Yoo, %-E.; Timmermans, P. B. M. W. M. Nonpeptide Angiotensin I1 Receptor Antagonists: The Discovery of a Series of N-(Biphenylmethy1)imidazolesas Potent, Orally Active Antihypertensives. J . Med. Chem. 1991, 34, 2525-2547.

JM9408676